Unverified50% confidenceFactExact time
Michael Truell stated that in SpaceX and Cursor production environments, Grok 4.7 uses only 5% more tokens for median requests and 20–30% more for p99 requests.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/25/2026
First Seen
Valid until: 12/24/2026
Sources
Related Entities
Related Claims
UnverifiedA 200K token context request requires approximately 25× the KV-Cache memory compared to an 8K token request.63% similarUnverifiedReal-world testing with the 284-billion-parameter DeepSeek V4 Flash shows token generation rising from 37 to 53 tokens/s (approximately 1.5x), and Prompt Processing jumping from 483 to 1,485 tokens/s (approximately three times).62% similarUnverifiedWhen running 4-bit DeepSeek R1 on the M3 Ultra, a 1,000-token context generates approximately 19 tokens per second using 392GB, dropping to approximately 15 tokens per second using 410GB at a 16,000-token context.60% similarUnverifiedIn a text generation test using an 8-billion-parameter model, the M5 Ultra achieved 188 tokens per second versus 129 tokens per second for the M3 Ultra, representing only 1.45x.60% similarUnverifiedGrok 4.7's token consumption of 81k exceeds that of models such as Fable 5.1.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/948936API
curl https://kongchang.com/api/v1/knowledge/claims/948936MCP
get_claim(id=948936)